Merged
Conversation
Move from a runtime `condition` on the ESRP steps to a YAML parse-time condition so that no usage of the `esrp*ConnectionName` variables exist when the esrp parameter is false. This means we can avoid the need to approve runs of this workflow in the internal secure environment when not accessing ESRP (for example, when debugging and testing the release process). Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
All other platforms use the package filename format: gcm(user)-$RUNTIME-$VERSION.$EXT But the Linux packages have been set to: gcm-$RUNTIME.$VERSION.$EXT Let's standardise on the `.` separator between runtime and version. Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
All other platforms use the package filename format: ``` gcm(user)-$RUNTIME-$VERSION.$EXT ``` But the Linux packages have been set to: ``` gcm-$RUNTIME.$VERSION.$EXT ``` Let's standardise on the `.` separator between runtime and version.
Move from a runtime `condition` on the ESRP steps to a YAML parse-time condition so that no usage of the `esrp*ConnectionName` variables exist when the esrp parameter is false. This means we can avoid the need to approve runs of this workflow in the internal secure environment when not accessing ESRP (for example, when debugging and testing the release process).
Mirroring #1811, add support to set default settings for Linux, via MDM tooling. For this platform we look for files in the `/etc/git-credential-manager/config.d` directory. Files are a simple format of `key=value\n`, where the `key` is a configuration value name from `docs/config.md`. This format prevents values from containing a `\n` line feed character, but this is unlikely to be an issue in practice.
Remove the cruft code signing summary file from the Windows payload zips. There's nothing interesting in this file anyway. Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
When building locally the gcm(user)-$RID-$VERSION.exe installer files were being written to the PayloadPath, rather than the OutputPath. Also, since we're now specifying a RID for all invocations of the project file, we should disable the appending of the RID to the OutputPath variable - we already have the RID as part of the PayloadPath (for keeping binaries separated by RID) and the resulting installer filenames also contain the RID anyway. Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
We assume the input paths given to the Windows layout.ps1 scripts do not have a trailing slash - we should make sure this is the case before proceeding and doing path-math with that. Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
Improve the handling of output paths and build artifacts in the Windows installer build pipeline. We accidentally started including some code signing summary report files in the output zip archives on Windows - let's delete those. Also I noticed that local builds were now outputting the Windows installers to the payload directory, rather than the expected output path. This is because now that we specify a RID we were getting the RID appended to the output path (which would then be the same as the payload path.. what a coincidence!). Finally let's be a bit more robust with the path-math in our layout.ps1 script - we should not assume if the path has the trailing slash or not. (Note that in MSBuild, the convention is that directory path variables should end with a slash.)
If there is no GCM configuration defaults directory on Linux, we had been inadvertently returning `null` for the setting value! Fix the issue by explictly returning `false` (no setting found) to `TryGetExternalDefault` rather than `true` (setting found). Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
If there is no GCM configuration defaults directory on Linux, we had been inadvertently returning `null` for the setting value! Fix the issue by explictly returning `false` (no setting found) to `TryGetExternalDefault` rather than `true` (setting found).
Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com>
dscho
approved these changes
Feb 4, 2026
Contributor
Author
|
Install from source on Ubuntu is a know issue. The version of the .NET SDK on that image has an issue. Unrelated to these changes. |
Contributor
Is the issue this? I'm seeing this on Gentoo with .NET SDK 9.0.111. Same with GCM 2.7.2. GCM 2.7.0 is fine. |
This comment was marked as spam.
This comment was marked as spam.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Changes: